Background: Failure to follow up abnormal test results is a significant safety concern in outpatient settings and often leads to patient harm and malpractice claims. Electronic health records (EHRs) can help ensure reliable delivery of abnormal test results, but they do not guarantee that this results in appropriate follow-up action. Our work in the Veterans Health Administration (VA) reveals that almost 8% of abnormal outpatient test results transmitted as EHR-based alerts lacked follow-up at 4 weeks. We subsequently found that follow-up of abnormal tests is influenced by multitude of technological factors (software/hardware) and non-technological factors (user behaviors, workflow, information load, policies and procedures, training and other organizational factors). Improving test result follow-up will require a better understanding of how follow-up processes fit within the complex socio-technical context of EHR-enabled health care. It is especially important to clarify how these contextual features influence the cognitive processes that are necessary to perceive, comprehend, and act on abnormal findings in a timely manner. Given that laboratory test result reporting is a component of Stage 2 meaningful use, further exploration of vulnerabilities in EHR-based test result follow-up is imperative. Objectives/Methods: We propose to apply human factors-based frameworks to understand system and cognitive vulnerabilities that affect EHR-based outpatient test result follow-up. To better define the context of clinical work that affects decision-making in this area, we will use a conceptual model that posits a set of eight socio-technical dimensions that must be considered in the real-world use of IT. Building on our prior work in the VA, our study settings include clinics affiliated with 3 non-VA institutions in order to improve generalizability. In Aim 1, we will identify the cognitive factors that affect test result follow-up processes in EHR-based health systems. We will conduct record reviews to identify recent abnormal test results with and without timely follow- up and conduct cognitive task analysis interviews with providers who ordered the tests. We will also assess the cognitive load of EHR-based alerts related to test results. In Aim 2, we will characterize the nature of clinical work required for individuals and teams to respond appropriately to abnormal test results in EHR-enabled outpatient settings. To map these processes at each site, we will collect qualitative data using rapid assessment techniques (structured observations, brief surveys, and key informant interviews). Our interpretation of these data will include consideration of how different socio-technical factors (e.g. EHR design, workflow, and organizational factors) interact and affect the cognitive work of test result follow-up. In Aim 3, we will conduct prospective risk assessments to characterize the particular work processes and features of the socio-technical context that are most vulnerable to failure within and across our study sites. This foundational work will lead to better understanding of the basic science of missed test results and will clarify targets for future interventions to improve follow-up of abnormal test results in EHR-enabled outpatient settings.